| (19) |
 |
|
(11) |
EP 0 270 160 B1 |
| (12) |
EUROPEAN PATENT SPECIFICATION |
| (45) |
Mention of the grant of the patent: |
|
19.01.1994 Bulletin 1994/03 |
| (22) |
Date of filing: 12.11.1987 |
|
| (51) |
International Patent Classification (IPC)5: H04L 27/22 |
|
| (54) |
Process and circuit for acquisition of carrier synchronism in coherent demodulators
Verfahren und Schaltung zur Rückgewinnung der Trägersynchronisierung in kohärenten
Demodulatoren
Procédé et circuit d'acquisition de synchronisme de la porteuse dans des modulateurs
cohérents
|
| (84) |
Designated Contracting States: |
|
DE ES FR GB GR IT SE |
| (30) |
Priority: |
05.12.1986 IT 2258486
|
| (43) |
Date of publication of application: |
|
08.06.1988 Bulletin 1988/23 |
| (73) |
Proprietor: SIEMENS TELECOMUNICAZIONI S.P.A. |
|
20060 Cassina de Pecchi (Milano) (IT) |
|
| (72) |
Inventors: |
|
- Lo Curto, Michelangelo
I-20038 Seregno (Milano) (IT)
- Ceroni, Ettore
I-24061 Albano S. Alessandro(Bergamo) (IT)
|
|
| |
|
| Note: Within nine months from the publication of the mention of the grant of the European
patent, any person may give notice to the European Patent Office of opposition to
the European patent
granted. Notice of opposition shall be filed in a written reasoned statement. It shall
not be deemed to
have been filed until the opposition fee has been paid. (Art. 99(1) European Patent
Convention).
|
[0001] The present invention relates to the field of digital signal transmissions and more
specifically to a process and related circuit for the acquisition of carrier synchronism
in coherent demodulators which receive at their input a PSK, or more general QAM,
modulated signal.
[0002] It is known that digital radio signal receivers include demodulators and that to
demodulate said signals, generally modulated by the Phase Shift Keying (PSK) technique,
it is necessary to reconstruct on the basis of the information contained in the received
signals a local carrier identical to the one that was suppressed in the transmitted
signals.
[0003] Among the methods used for reconstruction of the local carrier those most widely
used are derived from the Costas loop and to generate said carrier require the use
of a local voltage controlled oscillator (VCO) inserted in a phase-locked loop (PLL).
The VCO is controlled by means of an error signal function only of the phase-offset
existing between the local and the suppressed carrier.
[0004] A first limitation of the Costas loop technique and of all demodulation systems which
use an only-phase comparator for carrier locks is the limited capture range.
[0005] As is known the capture range of the phase-lock loop is given by the maximum frequency
difference (Δf) between the ouput signal frequency of the local oscillator in the
absence of a reference signal and the ouput signal frequency from the same local oscillator
which permits the loop to achieve phase lock in the presence of a reference signal.
[0006] Because of this limitation it is necessary to equip the phase-lock loop with search
devices to increase the capture range thereof as it will be better seen with reference
to Fig.1.
[0007] A second limitation of the Costas loop technique is that it makes possible false
locks, i.e. it can happen that the phase-lock loop finds stable points of equilibrium
even for particular frequency values of the local oscillator which do not correspond
to the input signal frequency, making impossible demodulation of said input signal.
[0008] In the article of G.L. Hedin, J.K. Holmes, W.C. Lindsey and K.T. Woo entitled "Theory
of False Lock in Costas Loops" published in IEEE Transactions on Communications, Vol.
COM-26, No. 1, January 1978 the authors examine some of the causes which bring about
false locks and show that the Δf
i frequencies at which they can occur have values which are a multiple of the symbol
frequency fs divided by the number of the modulation phases in accordance with the
formula:

, where n = 1,2,.... and, in case of Quadrature Phase-Shift Keying (QPSK), m = 4.
[0009] In the same article it is shown that the spectral distribution of the error signal
in lock condition is different depending on whether the lock is true or false. In
particular the components at multiple frequencies of the symbol frequency have different
amplitude. A first process is proposed to achieve correct lock consists first of detecting
the presence of false locks by measurement of the amplitude of the aforesaid components
and then forcing the phase-lock loop into the correct lock condition.
[0010] The proposed process is rather complicated, the hardware implementation is difficult
to accomplish and in particular it does not always assure correct operation.
[0011] In other articles there is introduced a parameter called 'False-Lock Margin' represented
by the ratio between the continuous component of the error signal under true lock
conditions and the continuous component of the error signal in the different possible
conditions of false lock, dependent in value on the transmission channel band. It
can be noted and experimentally confirmed that the false lock margin is extremely
reduced in QPSK systems, in which the first false lock takes place for


.
[0012] This suggests a second process for avoiding false locks which consists of limiting
the range of the local oscillator to values lower than ±fs/8. In this manner all false
locks would be avoided.
[0013] However this solution is not always practicable especially in small-capacity systems
for which the term fs/8 can be a very small fraction of the nominal carrier frequency.
Indeed, in these systems, because of the instability of the local oscillator, transmission
oscillator and local oscillators of the converters contained in the transceiver, the
nominal value of the input signal frequency has a certain indefiniteness which can
be greater than fs/8.
[0014] Hence this second process widely used for medium and large capacity systems is absolutely
unusable for small-capacity systems unless sophisticated and costly very high stability
oscillators are adopted.
[0015] A phase locked loop circuit, derived from the Costas loop with the addition of means
for extending the capture range of the PLL avoiding false locks, is disclosed in the
US patent US-A-4,338,574, inventors Fujita et all. The additional means consist of:
a delay line which opportunely delays a first error signal generated inside the loop;
a comparator which compares said first delayed error signal with the only-phase error
signal at the output from the Costas loop, obtaining a third error signal only depending
on the frequency offset existing between the local and the suppressed carrier. This
third only-frequency depending error signal is low-pass filtered and becomes a second
control signal for the VCO. Even if the circuit of Fujita overcomes the main drawbacks
of the previous demodulators, it shows however the drawback of generating two different
error signals, namely an only-phase and an only-frequency one.
[0016] A third process proposed for extending the capture range in QPSK demodulators derived
from the Costas loop is described in the article by F.D. Natali entitled "AFC Tracking
Algorithms" published in IEEE Transactions on Communications, Vol. COM-32 No. 8, August
1984 and which consists of utilizing a phase-frequency comparator.
[0017] The proposed solution appears to be rather complex in terms of hardware implementation,
requiring the use of linear multipliers and analog-to-digital converters difficult
to use in small-capacity systems where low cost and small size are of primary importance.
[0018] The object of the present invention is to overcome the aforesaid drawbacks and indicate
a process for acquisition of carrier synchronism in coherent demodulators which by
making use of simple and low-cost hardware implementation gives the phase-lock loop
of the demodulator an added frequency characteristic, obtaining only one error signal
simultaneously depending on both phase and frequency. Said added frequency characteristic
ensures locking between the local carrier and the carrier (suppressed) of the received
signals, avoiding false locks, and also ensures a sufficiently ample capture range.
[0019] To achieve said aims the object of the present invention is a process for the acquisition
of the carrier sinchronism in coherent demodulators, as better explained in the claim
1. Other object of the present invention is a circuit for the acquisition of carrier
synchronism, as better explained in the claim 5.
[0020] Other aims and benefits of the present invention will be made clear by the detailed
description given below and the annexed drawings given merely as an explanatory and
nonlimiting example wherein:
FIG. 1 shows a block diagram of a demodulator in accordance with the known art,
FIG. 2 shows a block diagram of a demodulator in accordance with the present invention,
FIG. 3 shows a first hardware implementation of a block of FIG. 2,
FIG. 4 shows a second hardware implementation of the same block of FIG. 2,
FIG. 5 shows some experimental voltage versus frequency output characteristics of
the demodulator, which is the object of the present invention, and
FIG. 6 shows some experimental voltage versus frequency output characteristics of
the demodulator, which is the object of the present invention, in the presence of
thermal noise.
[0021] FIG. 1 represents a Quadrature Phase-Shift Keying (QPSK) or Offset Quadrature Phase-Shift
Keying (O-QPSK) demodulator of the Costas loop type in which 1 indicates a power splitter
at the input of which is presented a modulated reception signal IF and the outputs
of which are connected to the first input of a mixer 2 and to the first input of a
mixer 3 respectively.
[0022] At the second input of the mixer 2 arrives the output signal of a voltage-controlled
local oscillator 4. Said signal, through a 90° phase shifter 5, also reaches the second
input of the mixer 3.
[0023] The output of the mixer 2 is connected to the input of a low-pass filter 6 the output
of which is connected to an input of an adder 8, to an input of a subtractor 9 and
to the input of a regenerative comparator 10. The output of the mixer 3 is connected
to the input of a low-pass filter 7 the output of which is connected to the second
input of the adder 8, to the second input of the subtractor 9 and to the input of
a regenerative comparator 11.
[0024] The output of the adder 8 is connected to the input of a regenerative comparator
12 while the output of the subtractor 9 is connected to the input of a regenerative
comparator 13. The output of the comparator 10 is connected to an input of a multiplier
14 and to the input D of a flip-flop 17 while the output of the comparator 11 is connected
to the other input of the multiplier 14 and to the input D of a flip-flop 18.
[0025] The outputs of the comparators 12 and 13 are connected to the two inputs of a multiplier
15.
[0026] The outputs of the two multipliers 14 and 15 are connected to the two inputs of a
multiplier 16 the output of which is connected to the input of a low-pass filter 19.
The output of the filter 19 is connected to the input of an amplifier 20 the output
of which is connected to the first input of an adder 30 to the second input of which
is connected the output of a block 21. The output of the adder 30 is connected to
the control input of the local oscillator 4.
[0027] As is known, in operation the intermediate frequency IF signal is split by the splitter
1 to a first input of each of the mixers 2 and 3. To the second input of the mixer
2 arrives the output signal of the local oscillator 4 while at the second input of
the mixer 3 arrives the output signal of the local oscillator 4 phase shifted 90°
by the phase shifter 5.
[0028] The demodulated signals at the outputs of the mixers 2 and 3 pass through the low-pass
filters 6 and 7, which serve primarily to reduce the noise present in the demodulated
signal band and are indicated by A₁ and B₁ respectively. Through the adder 8 and the
subtractor 9 the two signals A1 and B1 are added together and subtracted from each
other, generating the signals A1+B1 and A1-B1 respectively. The four signals A1, B1,
A1+B1 and A1-B1 through the comparators 10, 11, 12 and 13 are digitalized, generating
four streams of digitalized data respectively A2, B2, A2+B2 and A2-B2.
[0029] The signals A2 and B2 reach the input D of the samplers 17 and 18 respectively to
the clock input CK of which arrives a clock signal CP taken from the demodulated signal
A1 or B1 or from the received signal IF. The samplers 17 and 18 regenerate the signals
A2 and B2, reconstructing the data streams transmitted and indicated by A4 and B4
respectively.
[0030] The signals A2 and B2 are also multiplied together by means of the multiplier 14.
The signals A2+B2 and A2-B2 are multipled together by means of the multiplier 15.
The output signals from the multipliers 14 and 15 are further multiplied together
by the multiplier 16, the output signal of which is the error signal e(ϑ).
[0031] The signals A2, B2, A2+B2 and A2-B2 being digital on two levels, the multipliers
14, 15 and 16 are logical gates which accomplish the OR-Exclusive function.
[0032] The blocks enclosed by the broken line of FIG. 1 represent a phase comparator of
the Costas loop type the output signal of which e(ϑ) represents the phase error signal
for control of the local oscillator 4 and where 8 is the phase shifting angle between
the local carrier and the modulation carrier. The information on the modulation carrier
phase is then taken from the signals A1 and B1 and the function e(ϑ) is null only
when the local carrier phase coincides with that of the modulation carrier.
[0033] The low-pass filter 19 extracts the continuous component present in the error voltage
e(ϑ). Said component after being amplified by the amplifier 20 is applied to the first
input of the adder 30, to the second input of which arrives the output signal of the
block 21. The output signal of the adder 30 is applied to the control input of the
local oscillator 4 to allow it to achieve phase lock with the modulation carrier,
keeping this condition steady.
[0034] The block 21 is a known device for seeking lock which permits extending the capture
range and consists primarily of a triangle-waveform oscillator which, when the phase-lock
loop is unlocked, periodically varies the frequency of the voltage controlled local
oscillator 4 over its entire scanning range. When the lock has taken place the loop
reacts to counterbalance the output voltage of the triangle-waveform oscillator and
cancel out its effect.
[0035] FIG. 2, in which the same elements as in FIG. 1 are shown with the same numbers,
differs from FIG. 1 only by the addition of a block 22 having four inputs and four
outputs and by the absence of block 21 and adder 30.
[0036] The block 22 is placed between the outputs of the comparators 10, 11, 12 and 13 and
the inputs of the multipliers 14 and 15.
[0037] The blocks enclosed by the broken line of FIG. 2 represent a phase-frequency comparator
the output signal e(ϑ,f) of which represent the phase-frequency error signal for control
of the local oscillator 4.
[0038] The error signal e(ϑ,f) generated by the circuit in this configuration compared with
the known configuration of FIG. 1 is a function of the phase difference and of the
frequency difference between the locally generated carrier and the modulation carrier.
[0039] The function of the block 22 is to process the four signals A2, B2, A2+B2 and A2-B2
in such a manner that in its output signals A3, B3, A3+B3 and A3-B3 there is present
information on the carrier frequency (suppressed) of the signal received, i.e. said
block 22 gives to the phase-lock loop an added frequency characteristic and also permits
extension of the capture range.
[0040] FIG. 3 represents an embodiment of the circuit of the block 22 for a QPSK demodulator.
Within the block there is a known oscillator 23 which generates a signal the frequency
of which is indicated by 2fc. Said signal is sent to the three clock inputs CK of
three type-D flip-flops 24, 26 and 27. The output Q of flip-flop 24 is connected to
the input D of said flip-flop 24 to achieve the known configuraton of the divider
by two. Therefore, at the output Q of the flip-flop 24 there is present a clock signal
at frequency fc which together with the frequency 2fc serves for the sampling of the
input signals of the block 22. The output Q of the flip-flop 24 is connected to the
two clock inputs CK of two flip-flops 25 and 28.
[0041] The input D of the flip-flop 25 is connected to the output of the comparator 10 while
the output Q of said flip-flop is connected to the first input of the multiplier 14.
[0042] The input D of the flip-flop 28 is connected to the output of the comparator 11 while
the output Q of said flip-flop is connected to the second input of the multiplier
14.
[0043] The input D of the flip-flop 26 is connected to the output of the comparator 12 while
the output Q of said flip-flop is connected to the first input of the multiplier 15.
[0044] The input D of the flip-flop 27 is connected to the output of the comparator 13 while
the output Q of said flip-flop is connected to the second input of the multiplier
15.
[0045] FIG. 4 represents an embodiment of the circuit of the block 22 for an O-QPSK demodulator
and differs from FIG. 3 in that between the output of the oscillator 23 and the clock
input CK of the flip-flop 24 is inserted an invertor 29 and in that the clock signal
at the flip-flop 25 is not fc but fc, i.e. a phase shifted signal by one-half period
from fc.
[0046] In operation the circuit of FIG. 3 samples the signals A2 and B2 at frequency fc
and the sampling of signals A2+B2 and A2-B2 with double frequency, i.e. 2fc. The signals
A2+B2 and A2-B2 can be sampled with any frequency that is an even multiple of fc.
[0047] Operation of the circuit of FIG. 4 differs from that of FIG. 3 in that the signal
A3 is sampled at frequency fc, i.e. with a signal in phase opposition compared to
the signal with frequency fc. As is known, in an O-QPSK system the two demodulated
signals A1 and B1 are phase shifted with each other by a half period so that the respective
sampling signals must also be phase shifted by a half period.
[0048] In addition, for both solutions the frequency fc can be equal to or greater than
the symbol frequency fs of the signal received. In the first case there is synchronous
sampling and in the second case there is asynchronous sampling.
[0049] With synchronous sampling the oscillator 23 will clearly no longer be a free oscillator
but will have to be an oscillator controlled by the input signal so that it oscillates
at a frequency such that fc = fs. In addition, to avoid the sampling signals at frequency
fc or 2fc falling into the transition zone of signals A2, B2, A2+B2 and A2-B2 which
are zones devoid of information a phase shifter is required which will permit regulation
of the phase of the sampling signals.
[0050] With asynchronous sampling it has been found that the behaviour of the circuit depends
on the relationship between fc and fs. This behaviour is explained in FIGS. 5 and
6.
[0051] FIG. 5 represents the voltage versus frequency output characteristic experimentally
taken with an open loop of the phase-frequency comparator which is the object of the
present invention.
[0052] Specifically it represents the mean value of voltage e(ϑ,f) at the output of the
phase-frequency comparator, dependent in value on the frequency difference Δf between
the frequency of the local oscillator 4 and the frequency of the carrier (suppressed)
of the input signal, normalized with reference to the symbol frequency fs.
[0053] FIG. 5 shows three curves indicated with the letters a, b and c. Curve a represents
the output characteristic of a phase-only comparator accomplished in accordance with
the known art. Curves b and c were obtained from the phase-frequency comparator included
in the circuit for the acquisition of carrier synchronism which is the object of the
present invention for different values of fc/fs, specifically for

(synchronous sampling) and for

(asynchronous sampling).
[0054] Observing curve a it is seen that as freqency Δf varies, the output voltage of the
known comparator is always zero, i.e. there is no variation of the error signal e(ϑ)
with variations in the frequency.
[0055] Observing the curves b and c it is seen that as the frequency Δf varies, the error
signal e(ϑ,f) which drives the local oscillator for generation of the local carrier
at the exact frequency of the signal carrier varies, preventing false locks. Curve
c also shows a more extensive capture range than curve b. The capture range is represented
by the value Δf so that the error voltage e(ϑ,f) reverses its sign.
[0056] For an increase in the capture range there is a corresponding reduction of the slope
of the curve and hence a reduction in the gain of the phase-frequency comparator.
Choice of the sampling frequency will therefore depend on the specific characteristics
and performances required of the demodulator.
[0057] FIG. 6 shows curve c, which is the same as that of FIG. 5 obtained with

and in the absence of noise at the input of the demodulator (signal/noise ratio
S/N = ∞ ), and curve c1 which is also obtained with fc/fs = 3 with a signal/noise
ratio S/N of 4dB. The condition S/N = 4dB is the worst case so that the real curves
will stay in the area between c and c1 to assure excellent operation at all times.
[0058] As is known, the QAM modulation technique is derived from PSK modulation, having
in addition amplitude modulation. Consequently everything said for the QPSK modulator
applies to a QAM demodulator. In particular for recovery of the modulation carrier
the same phase-frequency comparator which is the object of the present invention is
applicable.
[0059] From the description given the advantages of the process and of the circuit for the
acquisition of carrier synchronism for coherent demodulators which are the object
of the present invention are clear. Specifically they are represented by the fact
that the demodulators produced in accordance with the proposed process are free from
false locks and have a very extensive capture range obtained without the help of a
supplementary search device. Furthermore, by merely varying the sampling frequency
fc it is possible to control within fairly extensive limits the capture range. Lastly,
the circuit is very simple and economical and being completely digital can be readily
accomplished by integrated circuit technology.
[0060] It is clear that numerous variants of the process and the circuit for acquisition
of carrier synchronism for coherent demodulators described as an example are possible
for those skilled in the art without thereby exceeding the scope of the claims.
1. Process for acquisition of carrier synchronism in coherent demodulators in which an
intermediate frequency receiving signal (IF), PSK or QAM modulated, is splitted (1)
into two equal signals, a first one being frequency multiplied by a locally generated
carrier obtaining a first demodulated signal (A1), a second one being frequency multiplied
by an in-quadrature locally generated carrier obtaining a second demodulated signal
(B1), the first and second demodulated signals being in turn added and subtracted
toghether obtaining a sum and a difference signals (A1+B1, A1-B1) which are digitalized
(A2+B2, A2-B2) and multiplied (15) obtaining a first multiplied signal, the first
and second demodulated signals (A1, B1) being digitalized (A2, B2) and multiplied
(14) obtaining a second multiplied signal, the first and second multiplied signals
being multiplied (16) obtaining an only-phase error signal (e(ϑ)), low-pass filtered
(19, 20) to control said locally generated carrier, characterized in that:
the sum and the difference digitalized signals (A2+B2, A2-B2) are sampled by a
first sampling signal with a first frequency (2fc) higher than the symbol frequency fs of said receiving signal (IF);
the first and second demodulated digitalized signals (A2, B2) are sampled by means
of a second sampling signal with a second frequency fc even sub-multiple, preferably 1/2, of said first frequency (2fc) and equal or higher than said symbol frequency fs;
the sampling of said: sum, difference, first and second demodulated digitalized
signals adding to said only-phase error signal (e(ϑ)) a further dependance (e(ϑ,f))
from the frequency offset (f) existing between said locally generated carrier and
the suppressed modulation carrier of said receiving signal (IF).
2. Process for acquisition of carrier synchronism in accordance with claim 1, characterized
in that both said first and second sampling frequencies (2fc, fc) can be increased or decreased in respect of said symbol frequency fs, in order to respectively extending or reducing the acquisition frequency range of
said carrier synchronism.
3. Process for acquisition of carrier synchronism in accordance with claim 1, characterized
in that said first and second sampling signals are appropriately shifted of a same
phase amount in comparison with said suppressed modulation carrier, avoiding, in case
of synchronous sampling with fc = fs, the sampling falling within transition zones of to-be-sampled digitalized signals
(A2, B2, A2+B2, A2-B2) devoid of information.
4. Process for acquisition of carrier synchronism in accordance with claim 1, characterized
in that said first demodulated digitalized signal (A2) is sampled by the second sampling
signal half-period shifted in comparison with said second sampling signal which samples
said second demodulated digitalized signal (B2), in order to correctly sample an intermediate
frequency receiving signal (IF) O-PSK modulated.
5. Circuit for acquisition of carrier synchronism in coherent demodulators comprising:
a power splitter (1) splitting an intermediate frequency receiving signal (IF), PSK
or QAM modulated, into two equal signals reaching a first input of a respective first
(2) and second (3) mixer whose second inputs are respectively reached by a local carrier
generated from a VCO (4) and by an in-quadrature local carrier obtained shifting said
local carrier by means of a quarter-wave shifter (5), the mixers output signals being
a first and a second demodulated signals (A1, B1) which are in turn added and subtracted
together by means of an adder (8) and a subtractor (9) circuit, obtaining a sum (A2+B2)
and a difference (A2-B2) signals which are digitalized by regenerative comparators
(12, 13) and multiplied (15) obtaining a first multiplied signal; said first and second
demodulated signals (A1, B1) being digitalized (A2, B2) and multiplied (14) obtaining
a second multiplied signal, the first and second multiplied signals being multiplied
(16) obtaining an only-phase error signal (e(ϑ)) which, after low-pass filtering (19,
20), reaches a control input of said VCO (4), characterized in that it further comprises:
oscillator means (23) for generating a first sampling signal with a first frequency
(2fc) higher than the symbol frequency fs of said receiving signal (IF);
frequency dividing means (24) to divide said first frequency (2fc) by an even number, preferably 2, obtaining a second sampling signal with a second
frequency fc equal or higher than the symbol frequency fs; and
a first and a second sampling circuits (25, 28), clocked by said second sampling
signal, for sampling the first and second demodulated digitalized signals (A2, B2);
a third and a fourth sampling circuits (26, 27), clocked by said first sampling
signal (2fc), are provided for sampling the sum and the difference digitalized signals (A2+B2,
A2-B2);
the sampling of said: sum, difference, first and second demodulated digitalized
signals adding to said only-phase error signal (e(ϑ)) a further dependance (e(ϑ,f))
from the frequency offset (f) existing between said local generated carrier and the
suppressed modulation carrier of said receiving signal (IF).
6. Circuit for acquisition of carrier synchronism in accordance with claim 5, characterized
in that said oscillator means (23) include means for increasing or decreasing the
value of said first frequency (2fc) of said first sampling signal in comparison with said symbol frequency fs, in order to respectively extend or reduce the acquisition frequency range of said
carrier synchronism.
7. Circuit for acquisition of carrier synchronism in accordance with claim 5, characterized
in that:
said first (26), second (27), third (25) and a fourth (28) sampling circuits are
respectively a first (26), second (27), third (25) and fourth (28) flip-flops;
in that said frequency dividing means is a fifth flip-flop (24) dividing said first
frequency (2fc) of said first sampling signal by two;
in that both said first (26) and second (27) flip-flops are clocked by said first
sampling signal; and
in that both said third (25) and fourth (28) flip-flops are clocked by a true output
signal of said fifth flip-flop, in order to correctly sample an intermediate frequency
receiving signal (IF) PSK or QAM modulated.
8. Circuit for acquisition of carrier synchronism in accordance with claim 5, characterized
in that:
said first (26), second (27), third (25) and a fourth (28) sampling circuits are
respectively a first (26), second (27), third (25) and fourth (28) flip-flops;
in that said frequency dividing means is a fifth flip-flop (24) dividing said first
frequency (2fc) of said first sampling signal by two;
in that said fifth flip-flop (24) is clocked by a first sampling signal inverted;
in that both said first (26) and second (27) flip-flops are clocked by said first
sampling signal;
in that said fourth flip-flop (28) is clocked by a non-inverted output signal of
said fifth flip-flop; and
in that said third flip-flop (25) is clocked by an inverted output signal of said
fifth flip-flop, in order to correctly sample an intermediate frequency receiving
signal (IF) O-PSK modulated.
9. Circuit for acquisition of carrier synchronism in accordance with claim 5, characterized
in that:
said oscillator means (23) comprise an oscillator controlled by said intermediate
frequency receiving signal (IF) for generating a first sampling signal whose first
frequency (2fc) is an exact even multiple of said symbol frequency fs; and
in that shifting means are provided to opportunely shift said first sampling signal
in comparison with said suppressed modulation carrier, avoiding the sampling falling
within transition zones of the to-be-sampled digitalized signals (A2, B2, A2+B2, A2-B2)
devoid of information.
1. Verfahren zur Rückgewinnung der Trägersynchronisierung in kohärenten Demodulatoren,
in denen ein empfangenes PSK- oder QAM-moduliertes ZF-Signal in zwei gleiche Signale
aufgespalten wird (1), wobei die Frequenz des ersten Signals mit einer örtlich erzeugten
Trägerfrequenz multipliziert wird, sodaß ein erstes demoduliertes Signal (A1) entsteht,
wobei die Frequenz des zweiten Signals mit einer örtlich erzeugten Quadratur-Trägerfrequenz
multipliziert wird, sodaß ein zweites demoduliertes Signal (B1) entsteht, wobei die
ersten und zweiten demodulierten Signale der Reihe nach miteinander addiert und voneinander
abgezogen werden, sodaß Summen- und Differenzsignale (A1+B1, A1-B1) entstehen, die
dann digitalisiert (A2+B2, A2-B2) und multipliziert werden (15), sodaß ein erstes
multipliziertes Signal entsteht, wobei die ersten und zweiten demodulierten Signale
(A1, B1) digitalisiert (A2, B2) und multipliziert werden (14), sodaß ein zweites multipliziertes
Signal entsteht, wobei die ersten und zweiten multiplizierten Signale miteinander
multipliziert werden (16), sodaß ein Phasenfehlersignal (e(ϑ)) entsteht, das über
einen Tiefpaßfilter lauft (19, 20), um den genannten örtlich erzeugten Träger steuern
zu können, dadurch gekennzeichnet, daß:
die Summen- und Differenz-Digitalsignale (A2+B2, A2-B2) durch ein erstes Abtastsignal
mit einer ersten Frequenz (2fc) abgetastet werden, die höher als die Symbolfrequenz fs des genannten Empfangssignals (ZF) ist;
die ersten und zweiten demodulierten Digitalsignale (A2, B2) durch ein zweites
Abtastsignal mit einer zweiten Frequenz fc abgetastet werden, die ein geradzahliger Teiler, vorzugsweise 1/2, der genannten
ersten Frequenz (2fc) ist, und zwar gleich oder höher als die genannte Symbolfrequenz fs;
die Abtastung der genannten Summen-, Differenz-, ersten und zweiten demodulierten
Digitalsignale dem genannten Phasenfehlersignal (e(ϑ)) eine weitere Abhängigkeit (e(ϑ,f))
von der Frequenzabweichung (f) hinzufügt, die zwischen dem örtlich erzeugten Träger
und dem unterdrückten Modulationsträger des genannten Empfangssignals (ZF) besteht.
2. Verfahren zur Rückgewinnung der Trägersynchronisierung nach Anspruch 1, dadurch gekennzeichnet,
daß die beiden genannten ersten und zweiten Abtastfrequenzen (2fc, fc) gegenüber der genannten Symbolfrequenz fs erhöht oder erniedrigt werden können, sodaß jeweils der Frequenzbereich der genannten
Trägergleichlaufanpassung erweitert oder reduziert werden kann.
3. Verfahren zur Rückgewinnung der Trägersynchronisierung nach Anspruch 1, dadurch gekennzeichnet,
daß die beiden genannten ersten und zweiten Abtastfrequenzen in geeigneter Weise gegenüber
dem genannten unterdrückten Modulationsträger um den gleichen Phasenbetrag verschoben
werden, sodaß im Falle von synchroner Abtastung mit fc = fs vermieden wird, daß die Abtastwerte in die Übergangszonen der noch abzutastenden
Digitalsignale (A2, B2, A2+B2, A2-B2) fallen, die keine Information enthalten.
4. Verfahren zur Rückgewinnung der Trägersynchronisierung nach Anspruch 1, dadurch gekennzeichnet,
daß das genannte erste demodulierte Digitalsignal (A2) durch die zweite, gegenüber
der genannten zweiten Abtastfrequenz verschobene Halbwelle des genannten zweiten Abtastsignals
abgetastet wird, die das genannte zweite demodulierte Digitalsignal (B2) abtastet,
sodaß ein empfangenes O-PSK-moduliertes ZF-Signal (ZF) einwandfrei abgetastet wird.
5. Schaltkreis zur Rückgewinnung der Trägersynchronisierung in kohärenten Demodulatoren,
der folgendes enthält: einen Leistungsteiler (1), der das empfangene PSK- oder QAM-modulierte
ZF-Signal (ZF) in zwei gleiche Signale aufteilt, die jeweils den ersten Eingang eines
ersten (2) und zweiten (3) Mischers erreichen, an dessen zweitem Eingang jeweils ein
von einem VCO (4) erzeugter örtlicher Träger und ein durch Verschiebung des genannten
örtlichen Trägers in einem Viertelwellenschieber (5) erzeugter örtlicher Quadraturträger
ansteht, wobei die Ausgangssignale des Mixers ein erstes und zweites demoduliertes
Signal (A1, B1) darstellen, die der Reihe nach in einem Summierschaltkreis (8) addiert
und in einem Subtraktionsschaltkreis (9) voneinander abgezogen werden, wodurch Summensignale
(A2+B2) und Differenzsignale (A2-B2) entstehen, die jeweils von einem regenerativen
Vergleicher (12,13) digitalisiert und multipliziert (15) werden, sodaß ein erstes
multipliziertes Signal entsteht; die genannten ersten und zweiten demodulierten Signale
(A1, B1) werden digitalisiert (A2, B2) und miteinander multipliziert (14), sodaß ein
zweites multipliziertes Signal entsteht, während die ersten und zweiten multiplizierten
Signale wiederum miteinander multipliziert (16) werden, sodaß ein Phasenfehlersignal
(e(ϑ)) entsteht, welches nach Filterung in einem Tiefpaßfilter (19,20) einen Steuereingang
des genannten VCO (4) erreicht, dadurch gekennzeichnet, daß er außerdem folgendes
enthält:
Einen Oszillatorschaltkreis (23) zur Erzeugung eines ersten Abtastsignals mit einr
ersten Frequenz (2fc), die höher ist als die Symbolfrequenz fs des genannten Empfangssignals (ZF);
Einen Frequenzteiler (24), der die genannte erste Frequenz (2fc) durch gerade Zahlen, vorzugsweise 2, teilt sodaß ein zweites Abtastsignal mit einer
zweiten Frequenz fc entsteht, die gleich oder höher ist als die Symbolfrequenz fs; und
ein erster und zweiter von dem genannten zweiten Abtastsignal getakteter Abtastschaltkreis
(25, 28), der die ersten und zweiten demodulierten digitalisierten Signale (A2, B2)
abtastet;
ein dritter und vierter von dem genannten ersten Abtastsignal (2fc) getakteter Abtastschaltkreis (26, 27), der die digitalisierten Summen- und Differenzsignale
(A2+B2, A2-B2) abtastet;
die Abtastung der genannten: Summen, Differenzen, ersten und zweiten demodulierten
Digitalsignale, die dem genannten Phasenfehlersignal (e(ϑ)) eine weitere Abhängigkeit
(e(ϑ,f)) von der zwischen dem genannten örtlich erzeugten Träger und dem unterdrückten
modulierten Träger des genannten Empfangssignals (ZF) bestehenden Frequenzverschiebung
(f) hinzufügt.
6. Schaltkreis zur Rückgewinnung der Trägersynchronisierung nach Anspruch 5, dadurch
gekennzeichnet, daß die Oszillatorschaltung (23) außerdem eine Schaltung zur Erhöhung
und Erniedrigung der genannten ersten Frequenz (2fc) des genannten ersten Abtastsignals gegenüber der genannten Symbolfrequenz (fs) beinhaltet, sodaß jeweils der Frequenzbereich der genannten Trägergleichlaufanpassung
erweitert oder reduziert werden kann.
7. Schaltkreis zur Rückgewinnung der Trägersynchronisierung nach Anspruch 5, dadurch
gekennzeichnet, daß:
die erste (26), zweite (27), dritte (25) und vierte (28) Abtastschaltung jeweils
ein erster (26), zweiter (27), dritter (25) und vierter (28) flip-flop sind;
die genannte Frequenzteilerschaltung ein fünfter flip-flop (24) ist, der die genannte
erste Frequenz (2fc) des ersten Abtastsignals durch 2 teilt;
die beiden genannten ersten (26) und zweiten (27) flip-flops von dem genannten
ersten Abtastsignal getaktet werden;
die genannten drittem (25) und vierten (28) flip-flops durch eine wirkliches Ausgangssignal
des fünften flip-flops getaktet werden, sodaß das empfangene PSK- oder QAM-modulierte
ZF-Signal (ZF) korrekt abgetastet wird.
8. Schaltkreis zur Rückgewinnung der Trägersynchronisierung nach Anspruch 5, dadurch
gekennzeichnet, daß:
die erste (26), zweite (27), dritte (25) und vierte (28) Abtastschaltung jeweils
ein erster (26), zweiter (27), dritter (25) und vierter (28) flip-flop sind;
die genannte Frequenzteilerschaltung ein fünfter flip-flop (24) ist, der die genannte
erste Frequenz (2fc) des ersten Abtastsignals durch 2 teilt;
der genannte fünfte flip-flop (24) durch ein erstes invertiertes Abtastsignal getaktet
wird;
die beiden genannten ersten (26) und zweiten (27) flip-flops von dem genannten
ersten Abtastsignal getaktet werden;
der genannte vierte flip-flop (28) durch ein nicht invertiertes Auysgangs-Signal
des genannten fünften flip-flops getaktet wird; und
der genannte dritte flip-flop (25) durch das invertierte Ausgangs-Signal des genannten
fünften flip-flops getaktet wird, sodaß ein empfangenes O-PSK-moduliertes ZF-Signal
(ZF) korrekt abgetastet wird.
9. Schaltkreis zur Rückgewinnung der Trägersynchronisierung nach Anspruch 5, dadurch
gekennzeichnet, daß:
die genannte Oszillatorschaltung (23) einen durch die genannte ZF-Empfangsfrequenz
(ZF) gesteuerten Oszillator beinhaltet, der ein erstes Abtastsignal erzeugt, dessen
erste Frequenz (2fc) ein exaktes geradzahliges Vielfaches der genannten Symbolfrequenz fs ist; und
ein Frequenzschieber vorhanden ist, der in geeigneter Weise das genannte erste
Abtastsignal gegenüber dem genannten unterdrückten Modulationsträger verschiebt, sodaß
vermieden wird, daß Abtastwerte in die Übergangszonen der noch abzutastenden Digitalsignale
(A2, B2, A2+B2, A2-B2) fallen, die keine Information enthalten.
1. Procédé d'acquisition de synchronisme de la porteuse dans des démodulateurs cohérents
dans lequel un signal de réception de fréquence intermédiaire (FI), modulé par déplacement
de phase ou modulé en amplitude en quadrature, est divisé (1) en deux signaux égaux,
un premier signal étant multiplié en fréquence par une porteuse générée localement
produisant un premier signal démodulé (A1), un deuxième signal étant multiplié en
fréquence par une porteuse générée localement en quadrature produisant un deuxième
signal démodulé (B1), les premier et deuxième signaux démodulés étant ajoutés l'un
à l'autre puis soustraits l'un de l'autre produisant un signal de somme et un signal
de différence (A1+B1, A1-B1) qui sont numérisés (A2+B2, A2-B2) et multipliés (15)
produisant un premier signal multiplié, les premier et deuxième signaux démodulés
(A1, B1) étant numérisés (A2, B2) et multipliés (14) produisant un deuxième signal
multiplié, les premier et deuxième signaux multipliés étant multipliés (16) produisant
un signal d'erreur de phase uniquement (E(Θ)), passés par un filtre passe-bas (19,
20) pour commander ladite porteuse générée localement, caractérisé en ce que:
les signaux numérisés de somme et de différence (A2+B2, A2-B2) sont échantillonnés
par un premier signal d'échantillonnage (2fc) supérieur à la fréquence symbole fs dudit signal de réception (FI);
les premier et deuxième signaux démodulés numérisés (A2, B2) sont échantillonnés
par un deuxième signal d'échantillonnage à un sous-multiple pair de la deuxième fréquence
fc, de préférence 1/2, de la dite première fréquence (2fc) et égale ou supérieure à ladite fréquence symbole fs;
l'échantillonnage desdits signaux de somme, différence, des premier et deuxième
signaux démodulés numérisés ajoutant audit signal d'erreur de phase uniquement (e(Θ))
une autre dépendance (e(Θ,f)) vis à vis du décalage de fréquence (f) existant entre
ladite porteuse générée localement et la porteuse de modulation supprimée dudit signal
de réception (FI).
2. Procédé d'acquisition de synchronisme de la porteuse selon la revendication 1, caractérisé
en ce que lesdites première et deuxième fréquences d'échantillonnage (2fc, fc) peuvent toutes les deux être augmentées ou diminuées relativement à ladite fréquence
symbole fs, afin d'étendre ou réduire respectivement la gamme de fréquences d'acquisition
dudit synchronisme de porteuse.
3. Procédé d'acquisition de synchronisme de la porteuse selon la revendication 1, caractérisé
en ce que lesdits premier et deuxième signaux d'échantillonnage sont déphasés de manière
appropriée d'une même valeur de phase en comparaison à ladite porteuse de modulation
supprimée, pour éviter, dans le cas de l'échantillonnage synchrone avec fc = fs, que l'échantillonnage ne tombe dans les zones de transition de signaux numérisés
à échantillonner (A2, B2, A2+B2, A2-B2) vides d'informations.
4. Procédé d'acquisition de synchronisme de la porteuse selon la revendication 1, caractérisé
en ce que ledit premier signal démodulé numérisé (A2) est échantillonné par le deuxième
signal d'échantillonnage déphasé d'une demi-période par rapport audit deuxième signal
d'échantillonnage qui échantillonne ledit deuxième signal démodulé numérisé (B2),
afin d'échantillonner correctement un signal de réception de fréquence intermédiaire
(FI), modulé par déplacement de phase quadrivalente décalé.
5. Circuit d'acquisition de synchronisme de la porteuse dans des démodulateurs cohérents
comprenant: un répartiteur de puissance (1) divisant un signal de réception de fréquence
intermédiaire (FI) modulé par déplacement de phase ou modulé en amplitude en quadrature,
en deux signaux égaux atteignant une première entrée d'un premier (2) et d'un deuxième
(3) mélangeurs respectivement dont les deuxième entrées sont respectivement atteintes
par une porteuse locale générée par un VCO (4) et par une porteuse locale en quadrature
obtenue en déphasant ladite porteuse locale au moyen d'un déphaseur quart-d'onde (5),
les signaux de sortie des mélangeurs étant un premier signal et un deuxième signal
démodulés (A1, B1) qui sont ajoutés l'un à l'autre puis soustraits l'un de l'autre
au moyen d'un circuit d'addition (8) et d'un circuit de soustraction (9), produisant
un signal de somme (A2+B2) et un signal de différence (A2-B2) qui sont numérisés par
des comparateurs régénérateurs (12, 13) et multipliés (15) pour produire un premier
signal multiplié; lesdits premier et deuxième signaux démodulés (A1, B1) étant numérisés
(A2, B2) et multipliés (14) produisant un deuxième signal multiplié, les premier et
deuxième signaux multipliés étant multipliés (16) produisant un signal d'erreur de
phase uniquement (e(Θ)) qui, après passage par un filtre passe-bas (19, 20), atteint
une entrée de commande dudit VCO (4), caractérisé en ce qu'en outre il comprend:
un moyen d'oscillation (23) pour générer un premier signal d'échantillonnage à
une première fréquence (2fc) supérieure à la fréquence symbole fs dudit signal de réception (FI);
un moyen de division de fréquence (24) pour diviser ladite première fréquence (2fc) par un nombre pair, de préférence 2, produisant un deuxième signal d'échantillonnage
à une deuxième fréquence fc égale ou supérieure à la fréquence symbole fs; et
un premier et un deuxième circuits d'échantillonnage (25, 28), cadencés par ledit
deuxième signal d'échantillonnage, pour échantillonner les premier et deuxième signaux
démodulés numérisés (A2, B2);
un troisième et quatrième circuits d'échantillonnage (26, 27), cadencés par ledit
premier signal d'échantillonnage (2fc) sont fournis pour échantillonner les signaux numérisés de somme et de différence
(A2+B2, A2-B2);
l'échantillonnage desdits signaux de somme, différence, des premier et deuxième
signaux démodulés numérisés ajoutant audit signal d'erreur de phase uniquement (e(Θ))
une autre dépendance (e(Θ,f)) vis à vis du décalage de fréquence (f) existant entre
ladite porteuse générée localement et la porteuse de modulation supprimée dudit signal
de réception (FI).
6. Circuit d'acquisition de synchronisme de la porteuse selon la revendication 5, caractérisé
en ce que ledit moyen d'oscillation (23) comprend un moyen pour augmenter ou diminuer
la valeur de ladite première fréquence (2fc) dudit premier signal d'échantillonnage en comparaison à ladite fréquence symbole
fs, afin d'étendre ou de réduire respectivement la gamme de fréquences d'acquisition
dudit synchronisme de porteuse.
7. Circuit d'acquisition de synchronisme de la porteuse selon la revendication 5, caractérisé
en ce que:
lesdits premier (26), deuxième (27), troisième (25) et un quatrième (28) circuits
d'échantillonnage sont respectivement une première (26), deuxième (27), troisième
(25) et quatrième (28) bascules;
en ce que ledit moyen de division de fréquence est une cinquième bascule (24) divisant
ladite première fréquence (2fc) dudit premier signal d'échantillonnage par deux;
en ce que lesdites première (26) et deuxième (27) bascules sont toutes les deux
cadencées par ledit premier signal d'échantillonnage; et
en ce que lesdites troisième (25) et quatrième (28) bascules sont toutes les deux
cadencées par un signal de sortie réelle de ladite cinquième bascule, afin d'échantillonner
correctement un signal de réception de fréquence intermédiaire (FI) modulé par déplacement
de phase ou modulé en amplitude quadrivalente.
8. Circuit d'acquisition de synchronisme de la porteuse selon la revendication 5, caractérisé
en ce que:
lesdits premier (26), deuxième (27), troisième (25) et un quatrième (28) circuits
d'échantillonnage sont respectivement une première (26), deuxième (27), troisième
(25) et quatrième (28) bascules;
en ce que ledit moyen de division de fréquence est une cinquième bascule (24) divisant
ladite première fréquence (2fc) dudit premier signal d'échantillonnage par deux;
en ce que ladite cinquième bascule (24) est cadencée par un premier signal d'échantillonnage
inversé;
en ce que lesdites première (26) et deuxième (27) bascules sont toutes les deux
cadencées par ledit premier signal d'échantillonnage; et
en ce que ladite quatrième bascule (28) est cadencée par un signal de sortie non
inversé de ladite cinquième bascule; et
en ce que ladite troisième bascule (25) est cadencée par un signal de sortie inversé
de ladite cinquième bascule, afin d'échantillonner correctement un signal de réception
de fréquence intermédiaire (FI) modulé par déplacement de phase décalé.
9. Circuit d'acquisition de synchronisme de la porteuse selon la revendication 5, caractérisé
en ce que:
ledit moyen d'oscillation (23) comprend un oscillateur commandé par ledit signal
de réception de fréquence intermédiaire (FI) pour générer un premier signal d'échantillonnage
dont la première fréquence (2fc) est un multiple pair exact de ladite fréquence symbole fs; et
en ce que des moyens de déphasage sont fournis pour déphaser comme il convient
ledit premier signal d'échantillonnage par rapport à ladite porteuse de modulation
supprimée, en évitant que l'échantillonnage ne tombe dans les zones de transition
des signaux numérisés à échantillonner (A2, B2, A2+B2, A2-B2) vides d'informations.